Geometric Understanding of Sketches
This work addresses the challenge of sketch interpretation for users like novices, experts, and artists, enabling downstream tasks such as robotic replication and 3D object completion, though it appears incremental by building on existing deep learning and graphics techniques.
The paper tackles the problem of enabling machines to geometrically understand sketches, presenting two methods: one for interpreting 2D line drawings as graphs and reconstructing them physically with a robot, achieving sub-pixel accuracy in vertex estimation, and another for completing contour-like sketches of 3D objects with illumination and texture, validated through a user interface and task-based exercises for artists.
Sketching is used as a ubiquitous tool of expression by novices and experts alike. In this thesis I explore two methods that help a system provide a geometric machine-understanding of sketches, and in-turn help a user accomplish a downstream task. The first work deals with interpretation of a 2D-line drawing as a graph structure, and also illustrates its effectiveness through its physical reconstruction by a robot. We setup a two-step pipeline to solve the problem. Formerly, we estimate the vertices of the graph with sub-pixel level accuracy. We achieve this using a combination of deep convolutional neural networks learned under a supervised setting for pixel-level estimation followed by the connected component analysis for clustering. Later we follow it up with a feedback-loop-based edge estimation method. To complement the graph-interpretation, we further perform data-interchange to a robot legible ASCII format, and thus teach a robot to replicate a line drawing. In the second work, we test the 3D-geometric understanding of a sketch-based system without explicit access to the information about 3D-geometry. The objective is to complete a contour-like sketch of a 3D-object, with illumination and texture information. We propose a data-driven approach to learn a conditional distribution modelled as deep convolutional neural networks to be trained under an adversarial setting; and we validate it against a human-in-the-loop. The method itself is further supported by synthetic data generation using constructive solid geometry following a standard graphics pipeline. In order to validate the efficacy of our method, we design a user-interface plugged into a popular sketch-based workflow, and setup a simple task-based exercise, for an artist. Thereafter, we also discover that form-exploration is an additional utility of our application.